Commercially Available Computer-Aided Detection System for Pulmonary Nodules on Thin-Section Images Using 64 Detectors-Row CT

摘要: Rationale and Objectives Most studies of computer-aided detection (CAD) for pulmonary nodules have focused on solid nodule detection. The aim this study was to evaluate the performance a commercially available CAD system in with or without ground-glass opacity (GGO) using 64-detector-row computed tomography compared visual interpretation. Materials Methods Computed tomographic examinations were performed 48 patients existing suspicious chest radiography. Three radiologists independently reported location pattern (GGO, solid, part solid) each candidate scans, assigned confidence score, then analyzed all scans system. A reference standard established by consensus panel different radiologists, who found 229 noncalcified diameters ≥ 4 mm. True-positive false-positive results levels used generate jackknife alternative free-response receiver-operating characteristic plots. Results The sensitivity GGO 3 (60%–80%) significantly higher than that (21%) (McNemar's test, P < .0001). For overall nodules, figure-of-merit values (P = .005–.04) analysis. part-solid greater those system, indicating no significant differences. Conclusion Radiologists are superior GGO, but can still play complementary role detecting GGO.